https://bugs.llvm.org/show_bug.cgi?id=47618
Bug ID: 47618
Summary: compile-time segfault with C code when using `-g`
Product: OpenMP
Version: unspecified
Hardware: PC
OS: Linux
Status: NEW
Severity: enhancement
Priority: P
Component: Clang Compiler Support
Assignee: unassignedclangb...@nondot.org
Reporter: bert...@anl.gov
CC: llvm-bugs@lists.llvm.org
Hello,
I see a compile-time segfault if I compile a simple C OpenMP offload code with
`-g` (see reproducer below). It will compile and run properly without `-g`, and
it will also compile and run properly if it's renamed as a C++ file (test.cpp)
and compiled with clang++. This is a minor issue, since it has so many
workarounds.
# Steps to reproduce:
```
$ cat test.c
#include "stdio.h"
int main()
{
int n = 0;
#pragma omp target map(from:n)
n = 9;
printf( "Expected: 9, Got: %d\n", n);
return 0;
}
$ clang -fopenmp -fopenmp-targets=nvptx64-nvidia-cuda -g test.c
clang-12: error: unable to execute command: Segmentation fault
clang-12: error: ptxas command failed due to signal (use -v to see invocation)
clang version 12.0.0
(/gpfs/jlse-fs0/users/yeluo/opt/llvm-clang/llvm-project/clang
66bcb14312a08b5d7e1197d23d748b2e23c4d852)
Target: x86_64-unknown-linux-gnu
Thread model: posix
InstalledDir: /soft/compilers/llvm/master-nightly/bin
clang-12: note: diagnostic msg: Error generating preprocessed source(s).
```
It works fine without `-g`, so looks like it has something to do with that:
```
$ clang -fopenmp -fopenmp-targets=nvptx64-nvidia-cuda test.c
(base) [bertoni@gpu00 tmp]$ ./a.out
Expected: 9, Got: 9
```
# Compiler version:
All the versions I tried fail:
```
$ clang -v
clang version 12.0.0
(/gpfs/jlse-fs0/users/yeluo/opt/llvm-clang/llvm-project/clang
66bcb14312a08b5d7e1197d23d748b2e23c4d852)
Target: x86_64-unknown-linux-gnu
Thread model: posix
InstalledDir: /soft/compilers/llvm/master-nightly/bin
Found candidate GCC installation:
/soft/compilers/gcc/7.4.0/linux-rhel7-x86_64/lib/gcc/x86_64-pc-linux-gnu/7.4.0
Selected GCC installation:
/soft/compilers/gcc/7.4.0/linux-rhel7-x86_64/lib/gcc/x86_64-pc-linux-gnu/7.4.0
Candidate multilib: .;@m64
Selected multilib: .;@m64
Found CUDA installation: /soft/compilers/cuda/cuda-10.0.130, version 10.0
```
also
```
$ clang -v
clang version 12.0.0
(/gpfs/jlse-fs0/users/yeluo/opt/llvm-clang/llvm-project/clang
6313f5561945930e9a5ec63cb187605ce741bb61)
Target: x86_64-unknown-linux-gnu
Thread model: posix
InstalledDir: /soft/compilers/llvm/master-latest/bin
Found candidate GCC installation:
/soft/compilers/gcc/7.4.0/linux-rhel7-x86_64/lib/gcc/x86_64-pc-linux-gnu/7.4.0
Selected GCC installation:
/soft/compilers/gcc/7.4.0/linux-rhel7-x86_64/lib/gcc/x86_64-pc-linux-gnu/7.4.0
Candidate multilib: .;@m64
Selected multilib: .;@m64
Found CUDA installation: /soft/compilers/cuda/cuda-10.0.130, version 10.0
```
# Platform:
x86 + Nvidia V100
Thanks,
Colleen
--
You are receiving this mail because:
You are on the CC list for the bug._______________________________________________
llvm-bugs mailing list
llvm-bugs@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
